Re: ReSpec.js and diff marking

Robin Berjon wrote:
> Hi Shane,
>
> On Apr 15, 2010, at 18:36 , Shane McCarron wrote:
>   
>> I introduced a feature recently, and I didn't really tell anyone about it because it is sort of a proof-of-concept thing.  Basically, if you tell ReSpec.js where to find a diffmarking tool, then ctrl-alt-shift-s menu will show a diffmark option.  This option will take the current document and the document that is considered 'previous' and diff mark them into a new browser window.  I find this REALLY handy when editing.
>>     
>
> I agree that this is a useful feature, thanks for adding it!
>
> Plugging into the ctrl-alt-shift menu is good, I think that we should retool it to make it the one go-to place for extras.
>
>   
>> Here's how you use it:
>>
>>   set the configuration option diffTool:             "http://www3.aptest.com/standards/htmldiff/htmldiff.pl",
>>     
>
> Ideally, for simplicity's sake, it would be nice if ReSpec came out of the box with a default URI for such diffs. I understand if you'd rather we didn't use the one at your site. There's one carefully hidden in datedspace at http://www.w3.org/2007/10/htmldiff (see http://esw.w3.org/HtmlDiff for a list).
>   
I dont mind hardcoding mine.  Mine is on a server with lots of spare 
cycles and I have fixed some bugs in it.  The one at that date space 
location won't work with ReSpec.js as is.

>   
>> If your previousURI is not in TR space... and mine usually are not because I take snapshots of editors drafts for the working group to review, then you should also set previousURI.  If you don't set that, then it will figure it out from the other settings.
>>     
>
> I think that there should be a separate option (previousDiffURI?) that the diff would look at first before falling back to previousURI and then to the generated previousURI. The reason for this is that previousURI can be used to override the generated previous URI (typically, if the shortname changed) and if you're using it it could get a little bit clumsy to switch it to a previous ED snapshot and then back to the value you actually intend to use when publishing. That's not a big concern though to be honest.
>   

Good idea.  I will make this change and the one above.